Files
seaweedFS/weed/plugin/worker/handler_registry.go
Chris Lu 587c24ec89 plugin worker: support job type categories (all, default, heavy) (#8547)
* plugin worker: add handler registry with job categories

Introduce a self-registration pattern for plugin worker job handlers.
Each handler can register itself via init() with a HandlerFactory that
declares its job type, category (default/heavy), CLI aliases, and a
builder function.

ResolveHandlerFactories accepts a mix of category names ("all",
"default", "heavy") and explicit job type names/aliases, returning the
matching factories. This enables workers to be configured by resource
profile rather than requiring explicit job type enumeration.

* plugin worker: register all handlers via init()

Each job handler now self-registers into the global handler registry
with its canonical job type, category, CLI aliases, and build function:

  - vacuum:              category=default
  - volume_balance:      category=default
  - admin_script:        category=default
  - erasure_coding:      category=heavy
  - iceberg_maintenance: category=heavy

Adding a new job type now only requires adding the init() call in the
handler file itself — no other files need to be touched.

* plugin worker: replace hardcoded job type switch with registry

Remove buildPluginWorkerHandler, parsePluginWorkerJobTypes, and
canonicalPluginWorkerJobType from worker_runtime.go. The simplified
buildPluginWorkerHandlers now delegates to
pluginworker.ResolveHandlerFactories, which resolves category names
("all", "default", "heavy") and explicit job type names/aliases.

The default job type is changed from an explicit list to "all", so new
handlers registered via init() are automatically picked up.

Update all tests to use the new API.

* plugin worker: update CLI help text for job categories

Update the -jobType flag description and command examples to document
category support (all, default, heavy) alongside explicit job type names.

* plugin worker: address review feedback

- Add CategoryAll constant; use typed constants in tokenAsCategory
- Pre-allocate result slice in ResolveHandlerFactories
- Add vacuum aliases (vol.vacuum, volume.vacuum)
- List alias examples (ec, balance, iceberg) in -jobType flag help
- Create handlers aggregator package for subpackage blank imports so
  new handler subpackages only need to be added in one place
- Make category tests relationship-based (subset/union checks) instead
  of asserting exact handler counts
- Add clarifying comments to worker_test.go and mini_plugin_test.go
  listing expected handler names next to count assertions

---------

Co-authored-by: Copilot <copilot@github.com>
2026-03-07 18:30:58 -08:00

131 lines
3.4 KiB
Go

package pluginworker
import (
"fmt"
"strings"
"sync"
"google.golang.org/grpc"
)
// JobCategory groups job types by resource profile so that workers can be
// configured with a category name instead of an explicit list of job types.
type JobCategory string
const (
CategoryAll JobCategory = "all" // pseudo-category matching every handler
CategoryDefault JobCategory = "default" // lightweight, safe for any worker
CategoryHeavy JobCategory = "heavy" // resource-intensive jobs
)
// HandlerFactory describes how to build a JobHandler for a single job type.
type HandlerFactory struct {
// JobType is the canonical job type string (e.g. "vacuum").
JobType string
// Category controls which category label selects this handler.
Category JobCategory
// Aliases are alternative CLI names that resolve to this job type
// (e.g. "ec" for "erasure_coding").
Aliases []string
// Build constructs the JobHandler.
Build func(opts HandlerBuildOptions) (JobHandler, error)
}
// HandlerBuildOptions carries parameters forwarded from the CLI to handler
// constructors.
type HandlerBuildOptions struct {
GrpcDialOption grpc.DialOption
MaxExecute int
WorkingDir string
}
var (
registryMu sync.Mutex
registry []HandlerFactory
)
// RegisterHandler adds a handler factory to the global registry.
// It is intended to be called from handler init() functions.
func RegisterHandler(f HandlerFactory) {
registryMu.Lock()
defer registryMu.Unlock()
registry = append(registry, f)
}
// ResolveHandlerFactories takes a comma-separated token list that can contain
// category names ("all", "default", "heavy") and/or explicit job type names
// (including aliases). It returns a deduplicated, ordered slice of factories.
func ResolveHandlerFactories(tokens string) ([]HandlerFactory, error) {
registryMu.Lock()
snapshot := make([]HandlerFactory, len(registry))
copy(snapshot, registry)
registryMu.Unlock()
parts := strings.Split(tokens, ",")
result := make([]HandlerFactory, 0, len(snapshot))
seen := make(map[string]bool)
for _, raw := range parts {
tok := strings.ToLower(strings.TrimSpace(raw))
if tok == "" {
continue
}
if cat, ok := tokenAsCategory(tok); ok {
for _, f := range snapshot {
if cat == CategoryAll || f.Category == cat {
if !seen[f.JobType] {
seen[f.JobType] = true
result = append(result, f)
}
}
}
continue
}
f, err := findFactory(snapshot, tok)
if err != nil {
return nil, err
}
if !seen[f.JobType] {
seen[f.JobType] = true
result = append(result, f)
}
}
if len(result) == 0 {
return nil, fmt.Errorf("no job types resolved from %q", tokens)
}
return result, nil
}
// tokenAsCategory returns the category and true when tok is a known category
// keyword. "all" is treated as a special pseudo-category that matches every
// registered handler.
func tokenAsCategory(tok string) (JobCategory, bool) {
switch tok {
case string(CategoryAll):
return CategoryAll, true
case string(CategoryDefault):
return CategoryDefault, true
case string(CategoryHeavy):
return CategoryHeavy, true
default:
return "", false
}
}
func findFactory(factories []HandlerFactory, tok string) (HandlerFactory, error) {
for _, f := range factories {
if strings.EqualFold(f.JobType, tok) {
return f, nil
}
for _, alias := range f.Aliases {
if strings.EqualFold(alias, tok) {
return f, nil
}
}
}
return HandlerFactory{}, fmt.Errorf("unknown job type %q", tok)
}